INDEX

Explanations

references to numerical and statistical information

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UnusedPrivate

-0.71

 jspb

-0.59

 Vikipedi

-0.58

Rhestr

-0.58

 препратки

-0.57

 ?></

-0.56

يكب

-0.56

=[]

-0.55

 content

-0.55

hood

-0.54

POSITIVE LOGITS

capsulation

0.54

 Morte

0.52

 robber

0.51

 robbers

0.51

prepareStatement

0.49

 avenger

0.48

 duplex

0.47

requisites

0.47

Decided

0.47

 soltar

0.47

Activations Density 0.205%

references to numerical and statistical information

No Comments

No Known Activations